Behavioral Study of Software-Defined Network Parameters Using Exploratory Data Analysis and Regression-Based Sensitivity Analysis

To provide a low-cost methodical way for inference-driven insight into the assessment of SDN operations, behavioral study key network parameters that predicate proper functioning and performance software-defined networks (SDNs) is presented to characterize their alterations or variations, given various emulated scenarios. It standard practice use simulation environments investigate characteristics SDNs, quantitatively qualitatively; hence, scenarios typify investigated in this paper. The studied analytically are jitter, response time throughput SDN. These most vital metrics operations according literature, they have been behaviorally following popular states: normal operating condition without any incidents on SDN, hypertext transfer protocol (HTTP) flooding, transmission control (TCP) user datagram (UDP) when subjected distributed denial-of-service (DDoS) attack. implemented primarily via univariate multivariate exploratory data analysis (EDA) visualize variations each scenarios, linear regression-based draw inferences sensitivity Experimental results indicate (i.e., latency throughput) vary as scenario changes DDoS attack all sensitive respective with some level interactions between them.